Prognostic Effects of Pretreatment Statuses at Diagnosis in Patients with Chronic Myeloid Leukemia: Results of the New Target Observational Study 1

Background: Pretreatment status assessment at diagnosis in patients with chronic myeloid leukemia in chronic phase (CML-CP) is more important for successful treatment in the tyrosine kinase inhibitors (TKIs) era. For instance, the Charlson comorbidity index (CCI) score is considered the useful tool for predicting overall survival (OS) in patients with CML treated with imatinib (IM) (Saussele S et al., 2015). However, the effect of CCI in patients with CML treated with second-generation TKI (2nd TKI) and clinical practice setting was unknown. This study aimed to evaluate the effect of pretreatment statuses, including CCI, on OS of participants of the New TARGET observational study 1 conducted by the Japanese Society of Hematology (JSH).

Methods: Patients newly diagnosed with CML-CP were registered to the New TARGET observational study 1 from April 2010 to March 2013. They were treated with TKIs (IM, nilotinib [NIL], or dasatinib [DAS]) after registration. Exclusion criteria were defined as follows: (1) accelerated phase/blast crisis (AP/BC) CML and (2) pretreatment with interferon-alfa, any TKIs, or hydroxyurea for more than 3 months, or allogeneic hematopoietic stem cell transplantation before registration. Other details of the protocol were previously reported (Kizaki et al., 2019). Patients were classified into CCI risk groups of 2, 3, and ≥4 for analysis. The New TARGET observational study 1 was supported by research fundings from Novartis Pharmaceuticals and Bristol-Myers Squibb to JSH. This subgroup analysis was approved by the institutional review board of the Hamamatsu University School of Medicine. The chi-square test was used to compare clinical characteristics for categoricaldata and the Wilcoxon rank-sum test for continuous data. OS was calculated using the Kaplan-Meier method and compared by the log-rank test. Gray's test was used to compare cumulative incidence curves. Cox proportional hazard analyses were performed to determine prognostic indicators of OS. The Wald test was used to assess the prognostic significance of a candidate variable. Statistical analyses were performed using EZR, a graphical user interface for R (Kanda Y, 2013).

Results: Among 506 enrolled patients, 475 with a median age of 56 years were assessable. The median follow-up period was 5.4 years. In total, 103 patients (21.7%) had various types of comorbidities, with diabetes mellitus, mild liver disease, peripheral vascular disease, myocardial infarction, renal disease, and peptic ulcer disease (7.3%, 4.1%, 3.2%, 2.6%, 2.2%, and 2.2%, respectively) as the most common. The lowest CCI score was 2 owing to CML. CCI scores were stratified as follows: 2, 372 patients (78%); 3, 74 patients (16%); and ≥4, 29 patients (6%). Higher CCI scores were significantly associated with older age (P <.001). Overall, 151 patients were treated with IM, 175 with NIL, and 149 with DAS. Patients with higher CCI scores tended to be treated with IM. Differences in cumulative incidence of complete cytogenetic response, major molecular response, and AP/BC were not statistically significant in the observational period according to CCI scores (P = 0.46, P = 0.58, and P = 0.63, respectively). No differences in the treatment response among the CCI risk scores were found in the IM and 2nd TKI cohorts. The OS of patients with CCI scores of 2, 3, and ≥4 were 94.3%, 89.3%, and 73.9%, respectively (P <.001). Similar results for OS were observed in IM and 2nd TKI cohorts. In the multivariate analysis including the CCI, performance status (PS), the presence of additional chromosomal abnormalities (ACAs), EUTOS long-term survival score (ELTS), age (≥65 years), and the choice of initial TKI treatment (IM vs. 2nd TKIs), the CCI score of ≥4 and PS of ≥2 were identified as adverse prognostic factors for OS (Wald test, P <.01 and P = 0.02). Hazard ratios for the CCI score of ≥4 (vs. score 2) and PS of ≥2 (vs. PS 0) were 5.94 (95% confidence interval [CI]: 1.94-18.2; P <.01) and 6.01 (95% CI: 1.67-21.6; P <.01), respectively. ACAs, ELTS, and older age did not significantly affect OS in this cohort including patients treated with 2nd TKI.

Conclusion: Our results demonstrated that comorbidities and poor PS at diagnosis were important predictive factors for OS, regardless of the type of TKI in patients with CML-CP. The risk-adopted management based on comorbidities and PS may be important to further improve the clinical outcomes of CML-CP.

View largeDownload slide

View largeDownload slide

Close modal