Early molecular kinetics to predict deep molecular response and inform frontline TKI selection in pediatric chronic myeloid leukemia: A real-world multicenter study.

Abstract

254 Background: The prognostic utility of adult-derived scoring systems and early molecular response (EMR, defined as BCR :: ABL1 level ≤10% at 3 months) in pediatric chronic myeloid leukemia (CML) requires further validation. This study aimed to evaluate the EUTOS Long-Term Survival (ELTS) score and 3-month molecular response in a real-world pediatric cohort and compare frontline imatinib (IM) versus second-generation tyrosine kinase inhibitors (2G-TKIs). Methods: This multicenter, retrospective study analyzed 248 pediatric CML patients in chronic phase from 13 Chinese hospitals. Molecular responses were assessed per ELN recommendations. The majority (89.9%) received first-line imatinib therapy. The impact of the ELTS risk, 3-month BCR :: ABL1 level, TKI switching, age, baseline bone marrow fibrosis, and spleen size on outcomes was evaluated using Kaplan-Meier and Cox regression analyses. Results: With a median follow-up of 45.3 months, the 5-year overall and progression-free survival rates were 98.9% and 96.2%, respectively. The EMR group had faster rates of achieving major molecular response (MMR: median time 9 vs. 24 months, p≤0.001) and deep molecular response (DMR: median time 18 vs. 50 months, p≤0.001). Cumulative incidences were significantly higher for the EMR group (12-month MMR: 71.4% vs. 21.2%; HR 2.50; p < 0.001; 24-month DMR: 64.7% vs. 27.4%; HR 3.26;p < 0.001). Multivariable analysis confirmed 3-month BCR :: ABL1 > 10% was independent predictors of longer time to MMR and DMR. Furthermore, the EMR attained group also demonstrated superior 5-year event-free survival (74.5% vs. 49.7%; HR 2.89, p = 0.0006) and failure-free survival (73.6% vs. 47.8%; HR 2.92, p = 0.004). Of the patients who remained on their front-line therapy, front-line 2G-TKIs (n = 18) led to a significantly shorter median time to DMR (median time15 vs. 24 months, p = 0.0067) and higher 24-month DMR rate (100% vs. 52.5%, p = 0.00666) compared to IM (n = 168), though 12-month MMR rates were similar. However, baseline ELTS risk stratification (low- vs. intermediate/high-risk) showed no prognostic value for PFS, EFS and FFS. Conclusions: In pediatric CML, early molecular kinetics at 3 months are strong predictors of long-term outcomes. When the treatment goal is rapid and deep molecular response (e.g., aiming for future TFR), frontline 2G-TKIs may offer a significant benefit over IM. This decision must be balanced against their distinct toxicity profiles and cost. The prognostic value of the ELTS score in pediatric CML needs further assessment.