Abstract
Background: Chronic phase chronic myeloid leukemia (CML-CP) has become a manageable disease for most patients treated with tyrosine kinase inhibitors (TKIs). However, all TKIs have broad spectrum of toxic effects, and have to be managed by cessation, reduction and supportive care. The objective of this study is to analyze the adverse events (AEs) with different TKIs used as initial therapy for CML and their impact on outcome.
Methods: We retrospectively evaluated a total of 494 patients with CML who received at least one TKI, imatinib, dasatinib, nilotinib and bosutinib in a practice setting between 2004 and 2014 at multicenter participating in the Hokkaido hematology study group.
Results: Of the 494 patients (315 males and 179 females), with a median age of 59.5 years (range 2-93), imatinib, dasatinib or nilotinib were prescribed as the first line TKI in 283 (62.3%), 109 (24%) and 102 (22.5%) patients, respectively. Disease status at primary diagnosis was composed of chronic phase (450), accelerated phase (21) and blastic phase (23). With a median follow-up of 4.7 years in patients with CML-CP, the 5-year overall survival (OS), event-free survival (EFS) were 94.5% and 92.3%, respectively. The patients with complication or organ dysfunction (61/450, 13.6%) and age >60 (227/450, 50.4%) at diagnosis had significantly inferior OS (p= 0.0089 and p= 0.0012). The patients achieved higher rates of major molecular response (MMR) at 6 and 12 months after initial treatment with dasatinib, nilotinib vs imatinib (41.5%, 42.6% vs 12.5% and 54.3%, 54.5% vs 41.5%, p<.0001 and p<.0001), but final MMR rates were similar in dasatinib, nilotinib vs imatinib (70.2%, 70.3% vs 63.9%, p=0.179). Moreover, there were no significant differences in EFS and OS for specific TKIs (p= 0.345 and p= 0.458). Of the 450 patients with CML-CP, 312 treatment modifications after the first line TKI treatment were carried out: 144 (46.2%) TKI changes or definitive discontinuations, 60 (19.2%) dose reductions, 36 (11.5%) temporary discontinuations and 72 (23.3%) dose reductions after temporary discontinuation. The main reasons for the 312 treatment modifications were 254 AEs (81.4%) and 41 failure or progression (13.1%). After initial TKI treatment, 272 (60.4%), 118 (26.2%), 37 (8.2%) and 23 (5.1%) patients had no, 1, 2 and 3 TKI changes, respectively. However, the number of TKI changes was not related to OS and EFS (p= 0.574 and p= 0.267). In the first line TKI treatment, grade I-II and III-IV AEs occurred in 185 (41.1%) and 123 (27.3%) patients. AEs resulting in treatment modifications occurred in 142 (55.5%) patients for imatinib, 53 (53.5%) for nilotinib and 59 (62.1%) for dasatinib. Grade III-IV AEs in the first line TKI treatment was significantly correlated to inferior OS and EFS as compared with grade 0-II AEs (p= 0.00612 and p= 0.0014). Multivariate analyses confirmed the fact that grade III-IV AEs significantly predicted for inferior EFS and OS, HR=3.311, 95% CI 1.34-8.175 (p= 0.0094) and HR=3.096, 95% CI 1.4560-6.587 (p= 0.0033), respectively.
Conclusions: Although long-term outcomes were similar in each TKI regardless of the first line TKI selection, severe AEs in the first line TKI treatment decreased survival rate of the patients with CML-CP. We need the personalized or some specialized treatment for elderly patients or patients with frailty. Early change of TKIs is recommended, when encountered with severe AEs of specific TKIs.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.